Although schizophrenia researchers have a rich history of identifying psychometric confounds in clinical studies, problems continue to plague some areas of the field because these confounds have been addressed using measurement technologies developed in the early 20th century. In functional magnetic resonance imaging (fMRI), for instance, links between brain activation and behavior elicited through working memory (WM) tasks used during scanning can be obscured by confounds stemming from traditional measurement technology. The central premise of this grant is that applying modern measurement theory to task development is critical for using fMRI to test advanced brain-behavior models of cognitive psychopathology. Accordingly, this project will focus on training and research in psychometric imaging. The award will provide the candidate with support to pursue career development objectives that complement his expertise in advanced measurement theory and technology with training in fMRI with an applied focus on schizophrenia. The first training objective is for the candidate to obtain knowledge and skills needed to conduct schizophrenia research. In Specific Aim 1 he will use this training to show that matching item difficulty to WM ability through neurocognitive computerized adaptive testing minimizes and equates standard errors of ability estimates between patients and controls. This aim will establish an innovative methodology for eliminating the differential reliability confound in schizophrenia research. The second training objective is for the candidate to obtain knowledge and skills needed to conduct fMRI research. In Specific Aim 2 he will use this training to compare blood oxygen level-dependent response in dorsolateral prefrontal cortex between schizophrenia and control participants in order to test inverted-U and neural inefficiency brain activation models of WM deficits in schizophrenia. This aim will formally test two widely assumed but unconfirmed theories of cognitive psychopathology in the disorder. The third training objective is for the candidate to develop analytical skills needed to study functional bran networks. In Specific Aim 3 he will use this training to map spatiotemporal activation patterns in fMRI data using state-space modeling in order to explore the theory that patients show more disorganized neural recruitment than controls. This aim will robustly test the brain-behavior theory positing that WM deficits in schizophrenia are associated with poorly organized patterns of neural recruitment-as manifested by an increased number of brain-states and by activation that is uncoupled from performance-using a well-defined and modern psychometric testing approach. The candidate's long-term goals are to provide psychometric imaging consultation to researchers studying schizophrenia and other brain-based disorders and to develop an R01 plan to study advanced neuroimaging and mathematical modeling of cognitive psychopathology in psychotic disorders. This line of research will contribute to public health by identifying potential neural targets for psychiatric treatments aimed at reducing neurocognitive deficits in psychosis.